Auflistung nach Autor:in "Bours, Patrick"
1 - 8 von 8
Treffer pro Seite
Sortieroptionen
- KonferenzbeitragContinuous authentication using mouse dynamics(BIOSIG 2013, 2013) Mondal, Soumik; Bours, PatrickIn this paper, we demonstrate a new way to perform continuous authentication using Mouse Dynamics as the behavioural biometric modality. In the proposed scheme, the user will be authenticated per mouse event performed on his/her system. We have used a publicly available mouse dynamics dataset and extracted per event features suitable for the proposed scheme. In this research, we have used the mouse dynamics data of 49 users and evaluated the system performance with 6 machine learning algorithms. In this approach, the genuine user has never been classified as an impostor throughout a full session whereas the average number of mouse actions an impostor could perform before detection is 94 from the best classification algorithm with a person based threshold.
- KonferenzbeitragDetecting Sexual Predatory Chats by Perturbed Data and Balanced Ensembles Effects(BIOSIG 2021 - Proceedings of the 20th International Conference of the Biometrics Special Interest Group, 2021) Borj, Parisa Rezaee; Raja, Kiran; Bours, PatrickSecuring the safety of the children on online platforms is critical to avoid the mishaps of them being abused for sexual favors, which usually happens through predatory conversations. A number of approaches have been proposed to analyze the content of the messages to identify predatory conversations. However, due to the non-availability of large-scale predatory data, the stateof-the-art works employ a standard dataset that has less than 10% predatory conversations. Dealing with such heavy class imbalance is a challenge to devise reliable predatory detection approaches. We present a new approach for dealing with class imbalance using a hybrid sampling and class re-distribution to obtain an augmented dataset. To further improve the diversity of classifiers and features in the ensembles, we also propose to perturb the data along with augmentation in an iterative manner. Through a set of experiments, we demonstrate an improvement of 3% over the best stateof-the-art approach and results in an F1-score of 0.99 and an Fβ of 0.94 from the proposed approach.
- KonferenzbeitragDoes context matter for the performance of continuous authentication biometric systems? an empirical study on mobile devices(BIOSIG 2015, 2015) Mondal, Soumik; Bours, PatrickIn this paper we will show that context has an influence on the performance of a continuous authentication system. When context is considered we notice that the performance of the system improves by a factor of approximately 3. Even when testing and training are not based on exactly the same task, but on a similar task, we see an improvement of the performance over a system where the context is not included. In fact, we proof that the performance of the system depends on which particular kind of task is used for the training.
- KonferenzbeitragGait recognition for children over a longer period(BIOSIG 2011 – Proceedings of the Biometrics Special Interest Group, 2011) Derawi, Mohammad Omar; Balisane, Hewa; Bours, Patrick; Ahmed, Waqar; Twigg, PeterIn this paper a comparative investigation into the effects of time on gait recognition in children's walking has been carried out. Gait recognition has attracted considerable interest recently; however very little work has been reported in the literature which is related to gait recognition in children. It has been suggested ([Kyr02])that the gait of children does not stabilize before they are 11 years old. In this papers we will provide arguments that support this suggestion. When looking at the performance of gait recognition, which serves as an indicator for the stability of gait, we found a relationship between performance improvement and aging of children. The gait of a group of children was measured twice with a 6 months period between the two measurements. Our analysis showed that the similarity between these two measurements is significantly lower than the similarity within each of the measurements. Finally we also report the effect of gender on performance of gait recognition.
- KonferenzbeitragThe influence of fingerprint image degradations on the performance of biometric system and quality assessment(Biosig 2016, 2016) Liu, Xinwei; Pedersen, Marius; Charrier, Christophe; Busch, Christoph; Bours, Patrick
- KonferenzbeitragA Novel Mobilephone Application Authentication Approach based on Accelerometer and Gyroscope Data(BIOSIG 2018 - Proceedings of the 17th International Conference of the Biometrics Special Interest Group, 2018) Li, Guoqiang; Bours, PatrickThe advent of mobile phones have changed our daily life. We are heavily relying on various applications installed on mobilephones to communicate with other, to share personal information with them, and to access our bank account, etc. However, the security measurement in terms of accessing these applications is either omitted or user-hostile because of the burden of memorizing the PIN or password. In order to relieve people from such burden, we explore the possibility of developing a mobilephone application authentication approach by analyzing the accelerometer and gyroscope data collected from the first few seconds when the user opens an application. By evaluating several proposed authentication approaches on a dataset collected from a real-life scenario, we achieve the best EER at 22.72% by only using the data collected from first 3 seconds. We think integrating the proposed non-intrusive authentication approach into the mobilephone application as an alternative for PIN/password can provide a more user-friendly authentication mechanism.
- KonferenzbeitragPredicted Templates: Learning-curve Based Template Projection for Keystroke Dynamics(BIOSIG 2018 - Proceedings of the 17th International Conference of the Biometrics Special Interest Group, 2018) Khodabakhsh, Ali; Haasnoot, Erwin; Bours, PatrickKeystroke Dynamics (KD) as a biometric modality can provide authentication tools in many real-life applications, virtually at zero-cost on the client side, due to the reliance of these techniques on existing hardware, and their low computational expense. One promising application is the use of KD as a second factor in password-based authentication. A downside of the existing modeling methods is the assumption of stationary behavior from the clients. However, it is expected that humans show improvements in performing a specific task following practice. In this study, we propose methods for utilization of learning models in predicting the future behavior of the clients, even with little enrollment data, and generate predicted behavioral models that can be used in different classifiers. In our experiments, the predicted templates show a reduction in the average equal-error-rate (EER) consistently across different classifiers a benchmark dataset. A reduction of 20% is achieved on the best classifier. Given fewer enrollment data, the performance gain was shown to reach above 30%. Furthermore, we show that blind detection of attacks is possible, solely relying on the global learning curve, with an EER of 16%.
- KonferenzbeitragSoft biometrics database: a benchmark for keystroke dynamics biometric systems(BIOSIG 2013, 2013) Idrus, Syed Zulkarnain Syed; Cherrier, Estelle; Rosenberger, Christophe; Bours, PatrickAmong all the existing biometric modalities, authentication systems based on keystroke dynamics are particularly interesting for usability reasons. Many researchers proposed in the last decades some algorithms to increase the efficiency of this biometric modality. Propose in this paper: a benchmark testing suite composed of a database containing multiple data (keystroke dynamics templates, soft biometric traits . . . ), which will be made available for the research community and a software that is already available for the scientific community for the evaluation of keystroke dynamics based systems. We also built the proposed biometric database on soft biometric traits for keystroke dynamics to suit the experiment. 110 people had voluntarily participated and gave their soft biometrics data i.e. the way of typing, gender, age and handedness.